The Horse in Motion: Two Frames With No Hoof on the Track

In frames 2 and 3 of this card, not one of Sallie Gardner’s four hooves touches the track. Look at the legs in those two frames. They are not reaching out in front and stretching out behind. They are folded up under the belly, with the hooves bunched close together.

The card is one of Eadweard Muybridge’s 1878 series titled The Horse in Motion. Its printed caption says the mare was “running at a 1.40 gait over the Palo Alto track, 19th June, 1878,” ridden by G. Domm. So, in this series, a galloping horse does leave the ground. It does so in a pose unlike the stretched-out horses of the racing print further down this page, and frame 12 is not part of the gallop at all: in it the mare appears to stand still.

Which frames show the mare off the ground?

Run your eye along the top row. In frame 1 one foreleg is planted under the shoulder. In frames 2 and 3 every hoof sits above the line of the track, and the legs are tucked in so tight they almost form a ring. In frame 4 a hind leg reaches down and appears to meet the track again.

Frames 2 and 3 are the only ones on the card that show the mare unsupported. Every other frame from 4 to 11 is read as one or two hooves on the track, three of them as probable. The count below comes from looking at each frame of the high-resolution scan, one by one.

Open the diagram full size

These are readings, not measurements. The horse is a dark silhouette against a white wall, so left and right legs cannot be told apart, and three frames are marked probable. Muybridge had the same trouble with his early pictures. In his 1893 book he wrote that “it was difficult to distinguish, except by inference, the right feet from the left.”

Download the hoof count, frame by frame (CSV) · source and license notes

Eleven frames, one stride: the card as a loop

The card says its photographs “illustrate consecutive positions assumed during a single stride of the mare.” Frames 1 to 11 do follow on from one another, and frame 11 runs back into frame 1 without an obvious jump. Frame 12 does not belong. Put it into the loop and the galloping mare would stop dead for one frame and then race on.

So the loop below uses frames 1 to 11 only, played twice. The numbered scale along the top of each photograph is trimmed off, because its numbers change from frame to frame.

Sallie Gardner at a gallop, frames 1 to 11 twice

Every frame, side by side

1. 1card frame 1
2. 2card frame 2: no hoof on the track
3. 3card frame 3: no hoof on the track
4. 4
5. 5
6. 6
7. 7
8. 8
9. 9
10. 10
11. 11
12. 12
13. 13card frame 2: no hoof on the track
14. 14card frame 3: no hoof on the track
15. 15
16. 16
17. 17
18. 18
19. 19
20. 20
21. 21
22. 22

The card gives the time between pictures as “about the twenty-fifth part of a second.” At that pace the 11 positions would pass in under half a second. The loop runs at 10 frames per second instead, two and a half times slower, so frames 2 and 3 stay on screen long enough to see. At full speed they would fill only about 0.08 second: 2 frames at 25 a second.

The background behaves strangely when it moves. The card says the negatives were made at intervals of 27 inches, and the painted lines on the wall behind the track were also 27 inches apart. The numbers along the top of the photographs go up by one from each frame to the next, so each picture was taken about one line further along than the last. The lines barely shift, the mare stays in about the same place, and the legs do all the moving.

Spacing between pictures does the same work in bouncing ball animation, where even gaps and growing gaps tell the eye two different stories.

What the card itself says it measured

The small type at the foot of the card explains how the pictures were made. It calls the whole thing an “Automatic Electro-Photograph,” and it notes that a “Patent for apparatus” had been applied for. Then it gives the numbers:

• the negatives were made “at intervals of twenty-seven inches of distance, and about the twenty-fifth part of a second of time”
• the vertical lines on the wall were 27 inches apart, and the horizontal lines “represent elevations of four inches each”
• the negatives “were each exposed during the two-thousandth part of a second, and are absolutely ‘untouched’”
• the card is marked “Copyright, 1878, by Muybridge” and printed for “Morse’s Gallery, 417 Montgomery St., San Francisco”

The card does not say how a camera knew when to fire. Scientific American, writing about another set in the same year, said only that Muybridge “employed a series of cameras, operated by electricity,” placed to fix “the several phases in the continuous action of the horse while making one stride.” That is as far as these two sources go, so this page leaves the trigger at that.

Checking the card’s numbers

Some of the card’s numbers can be checked against each other. Read “a 1.40 gait” as a mile in 1 minute 40 seconds, the same pace Muybridge’s 1893 book gives in words for a galloping Kentucky horse, “a mile in a hundred seconds.” That is 52.8 feet per second. At that speed, 27 inches takes about 0.043 second, or one twenty-third of a second, close to the card’s “about the twenty-fifth part.”

Frames 1 to 11 sit 10 intervals apart. At 27 inches an interval, that is 270 inches, or 22.5 feet of track, covered in about 0.4 second at the card’s stated interval. For comparison, the trotter in Scientific American had a stride “over 18 feet” at his own pace, a different horse and a different gait.

The exposure figure is harder to test. Nothing on the card shows how one two-thousandth of a second was measured, so this page treats it as a claim printed on Muybridge’s card. The sharp legs in most frames fit a very short exposure. They do not prove the number.

How Scientific American saw the series in 1878

Scientific American put Muybridge’s horses on its front page on October 19, 1878. These are engravings, not photographs. The horse is a different one: the trotter Abe Edgington, shown walking (A to F) and trotting (1 to 12). The magazine said it had dropped “the driver and his sulky” and had the pictures “enlarged and skillfully engraved.”

The article on page 241 states that “the exposure for each negative was about the two thousandth part of a second.” That figure, per Scientific American, belongs to the Abe Edgington set. It matches the number printed on the Sallie Gardner card, but neither source shows the measurement.

The magazine counted hooves the same way this page does. The walking horse, it said, “always has two feet on the ground, and, for a brief interval in each stride, three feet.” The trotting horse leaves the ground too. “In Figs. 4 and 5, and again in 9 and 10,” the article says, “the horse is entirely off the ground, literally flying through the air.”

Then it turned to painters. At first, it predicted, an artist would say of many positions “that there is absolutely no ‘motion’ at all in them.”

Six series of cards, and a zoetrope

The same page says that “six series of cards have been published, with from eight to twelve positions each,” of trotting, cantering, running and walking horses. They could be bought from Muybridge at 417 Montgomery St., San Francisco. The Sallie Gardner card carries that address too, under the name “Morse’s Gallery.”

The article ends with a suggestion: mount the photographs “on strips for use in the zoetrope.” A zoetrope was a slotted drum that showed a ring of drawings one after another. Put photographs in place of drawings, and the strides on the cards would move again.

The stretched-out gallop of the racing prints

Here is a racing print from five years before the card. In the Currier & Ives lithograph of Tom Bowling winning the Jerome Stakes on October 4, 1873, all five horses sail over the track. Each one has its forelegs thrust out in front and its hind legs stretched out behind, like a rocking horse.

No frame of the Sallie Gardner card shows that pose in the air. The frames where the legs are stretched farthest, 7 and 8, have hooves on the track. The frames in the air, 2 and 3, have the legs folded in.

Muybridge had a name for the pose. Describing a greyhound in flight, his 1893 book compares its “elongated body and extended legs” to “the conventional galloping horse.” By his account the dog does fly that way. For the horse, his gallop analysis found only one phase without support, and it is not that one.

What Muybridge said the photographs settled

In his own account, the question that started it all was not about galloping. Writing in 1893, Muybridge said that in 1872 a “particular dispute” arose in San Francisco about “the possibility of a horse having all of his feet free of contact with the ground at the same instant, while trotting.” His first answer came from about a dozen negatives of the trotter Occident, made with a single camera at Sacramento in May 1872.

Six years later the camera line at Palo Alto gave whole strides. The book says results showing horses “walking, trotting, galloping, &c., were published in 1878, with the title of ‘The Horse in Motion,’” and that copies went to the Library of Congress the same year.

For the gallop, the 1893 book is specific. It lists the order in which the feet carry the horse, then adds that “the only phase in which he has been discovered without support is one when the legs are flexed under the body.” It also allows that a second moment in the air, between the lifting of one fore foot and the descent of the other, is probable but “has not yet been demonstrated.”

The tracings above come from his later photographs, so they are another horse on another day. Even so, phase 11 shows the same tucked pose as frames 2 and 3.

Off the card and onto a wall

Muybridge also took the pictures off the card. He wrote that at the Royal Institution in 1882 he used his Zoöpraxiscope and a lantern to project “a synthesis of many of the actions he had photographed.” That is his own account of the event.

A film copy such as The Great Train Robbery also shows single pictures in a row, but there they come off one camera at a steady rate. Muybridge’s came from a line of cameras, one picture each. A stereograph works the other way round: two lenses side by side catch one instant from two points of view, while Muybridge’s pictures caught many instants, each from about 27 inches further along the track than the one before.